All animal-related experiments were approved by the Institutional Animal Care and Use Committee of Chinese People's Liberation Army General Hospital [approval no. SCXK(JING)2019-0010; Beijing, China]. In addition, it is confirmed that animals were anesthetized and sacrificed using acceptable methods and techniques.

PMCs were prepared and cultured following the protocols described previously (21). Briefly, three 5-day-old neonatal C57BL/6 mice that had been euthanized with CO₂ inhalation (30% volume/min for at least 50 min) were purchased from SPEF (Beijing) Biotechnology Co., Ltd. Subsequently, the knee cartilage was isolated and digested using collagenase D solution at 0.5 mg/ml overnight at 37˚C (MilliporeSigma) to obtain PMCs. PMCs were cultured in DMEM/F12 (HyClone; Cytiva) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 0.1 mg/ml streptomycin (Gibco; Thermo Fisher Scientific, Inc.) in a culture dish at a density of 8x10³ cells/cm² under standard conditions (37˚C, 5% CO₂). After achieving confluence by days 6-7 of culture, PMCs were harvested using 0.25% Trypsin-EDTA (Gibco; Thermo Fisher Scientific, Inc.) and passaged. Passage two cells were used for all experiments.

PMCs were seeded into 96-well plates (5,000 cells/well) and then subjected to the selective treatments in sequential order: i) GCK (Chengdu Zhibiao Pure Biotechnology Co., Ltd.) pre-treatment (10 or 50 µM) for 24 h; ii) TNF-α (20 ng/ml; Beyotime Institute of Biotechnology) treatment for 12 h; and iii) GCK post-treatment (10 or 50 µM) for 12 h. Cells that received DMSO served as a negative control. Three sets of experiment groups were designed: Treatment i alone; treatment ii alone; and treatment i, ii and iii in combination. To check the influence of GCK on the viability of PMCs, cells in 96-well plates (5,000 cells/well) were cultured in the presence of different concentrations of GCK (0, 10, 20, 50, 100 and 150 µM) for 48 h. The cell viability of PMCs was then measured using a Cell Counting Kit-8 (CCK-8) assay (Dojindo Molecular Technologies, Inc.) following the manufacturer's instructions. In brief, 10 µl of CCK-8 solution was added to each well and incubated for 2 h. The absorbance at 450 nm was measured using an ELISA reader (Hangzhou Lianke Biotechnology Co., Ltd.).

Commercial ELISA kits were used to determine the concentration of IL-6 (cat. no. PI326; Beyotime Institute of Biotechnology), MMP13 (cat. no. YB-MMP13-Mu; Shanghai Yubo Biotechnology Co., Ltd.), A Disintegrin and Metalloproteinase with Thrombospondin Motifs 5 (ADAMTS5) (cat. no. LS-F32114; LifeSpan BioSciences, Inc.), MMP3 (cat. no. YB-MMP3-Mu; Shanghai Yubo Biotechnology Co., Ltd.), aggrecan (cat. no. YB-AGC-Mu; Shanghai Yubo Biotechnology Co., Ltd.) and collagen II (cat. no. YB-PIINP-Mu; Shanghai Yubo Biotechnology Co., Ltd.) in the culture supernatants after the aforementioned treatments, following the manufacturer's protocols.

PMCs (1x10⁵) were cultured on glass coverslips in 24-well plates and then, in sequential order, were treated as follows: i) Pre-treated with GCK (50 µM) for 24 h; ii) treated with TNF-α (20 ng/ml) for 12 h; and iii) post-treated with GCK (50 µM) for another 12 h. Cells that received DMSO treatment served as a negative control. Three sets of experiment groups were designed: Treatment i alone; treatment ii alone; and treatment i, ii and iii in combination. Following treatment, the cells were fixed in 4% paraformaldehyde for 15 min at room temperature, treated with 0.1% Triton X-100 and blocked using 3% bovine serum albumin (Gibco; Thermo Fisher Scientific, Inc.) for 1 h at room temperature. The coverslips were then incubated with anti-MMP13 (1:200; cat. no. 18165-1-AP; Wuhan Sanying Biotechnology) or anti-collagen II (1:200; cat. no. ab34712; Abcam) at 4˚C overnight. After washing three times with PBS (5 min/time) at room temperature, the coverslips were incubated with a secondary antibody (1:1,000; CoraLite488-conjugated anti-rabbit; cat. no. SA00013-2; Wuhan Sanying Biotechnology) at room temperature for 2 h. The nucleus was stained using DAPI (300 nM; Thermo Fisher Scientific, Inc.) for 5 min at room temperature in the dark.

For TUNEL staining, the fixation and Triton X-100 treatments were the same as aforementioned. A One Step TUNEL Apoptosis Assay Kit (cat. no. C1086, Beyotime Institute of Biotechnology) was used. TUNEL detection solution (50 µl) was added to the sample and incubated at 37˚C in the dark for 60 min. After washing two times with PBS (5 min/time) at room temperature, the fluorescence was detected. TUNEL-positive cells in five random views were quantified by manual counting. Fluorescent images were captured using a confocal microscope under x20 or x40 magnification (Leica Microsystems GmbH).

Pathogen-free WT C57BL/6J male mice (n=24; 2-month-old; body weight, 25-30 g) were obtained from SPEF (Beijing) Biotechnology Co., Ltd. The mice were housed in a temperature-controlled environment (temperature, 25±2˚C; relative humidity, 45-60%) with a 12-h light/dark cycle and received food and water ad libitum.

Before the experimental surgery, mice were anesthetized with 250 mg/kg intraperitoneal tribromoethanol (Avertin; Sigma-Aldrich; Merck KGaA). To prepare the 100% Avertin stock solution, 10 g tribromoethanol was added into a centrifuge tube with 10 ml tertiary amyl alcohol. The tube was then shaken in hand until the tribromoethanol was completely dissolved. The solution was then filtered using a 0.22-µm filter membrane. A working solution of 2.5% Avertin was prepared by diluting the 100% Avertin stock solution to 2.5% (1:40) with 0.9% NaCl. The working solution was stored at 4˚C in the dark and used within 2 weeks.

Experimental osteoarthritis was induced in the mice through transection of the medial meniscotibial ligament and medial collateral ligament in the right knee [destabilization of the medial meniscus (DMM) mice]. The left knees not subjected to surgery in the DMM alone group and was used as the control (sham) for both DMM and DMM + GCK groups. According to previous studies, 40 mg/kg GCK was sufficient to exert anti-inflammatory effects in both rat and mouse models of induced arthritis (22,23). The day after the surgery, the mice were fed for 8 weeks either with a control diet (normal diet) or with diets supplemented with GCK (40 mg/kg) (biological replicates, n=12 per group). Mice were euthanized with CO₂ inhalation (30% volume/min for at least 10 min) after 8 weeks of surgery. Euthanasia using CO₂ was conducted following the AVMA Guidelines for the Euthanasia of Animals (2020 edition). The mice were left in the CO₂ environment until the cessation of breathing and heartbeat, when fully dilated pupils were observed. Knee joints were then collected for histological analysis.

IHC, Safranin O-Fast Green and H&E staining were conducted in paraffin-embedded tissues using the BOND-III Automated IHC Stainer (Leica Microsystems GmbH). Knee joints were fixed in neutral buffered formalin at a concentration of 10%, at room temperature for 16 h. Dehydration was performed in graded alcohols (70, 95 and 100% ethanol). Next, the tissues were cleared in a clearing agent (xylene substitute) to remove alcohol and prepare them for infiltration with paraffin wax. The tissues were then infiltrated with liquid paraffin wax at 56-60˚C for 6 h until they became fully embedded. After that, the tissues were transferred into fresh liquid paraffin wax and allowed to cool and solidify. Sections (5 µm) were set in the Stainer. Antigens were retrieved by heating the tissue sections in a BOND Epitope Retrieval ER2 Solution contained in the BOND IHC Polymer Detection Kit (cat. no. DS9800) for 20 min at 100˚C. Tissues sections were then subjected to peroxide blocking using the hydrogen peroxide included in the BOND IHC Polymer Detection Kit (cat. no. DS9800) for 5 min. The following primary antibodies were used: Anti-collagen II (1:600; cat. no. 28459-1-AP; Wuhan Sanying Biotechnology), anti-MMP13 (1:250; cat. no. 18165-1-AP; Wuhan Sanying Biotechnology), anti-NLR family pyrin domain-containing 3 (anti-NLRP3; 1:200; cat. no. 19771-1-AP; Wuhan Sanying Biotechnology) and anti-Gasdermin D-N terminal (anti-GSDMD-NT; 1:800; cat. no. 36425; Cell Signaling Technology, Inc.). The tissue sections were incubated with the diluted primary antibodies for 30 min at room temperature and then washed three times with BOND Wash Solution (2 min/time). Next, the sections were incubated with an HRP-conjugated secondary antibody (1:2,000) contained in the BOND IHC Polymer Detection Kit (cat. no. DS9800) for 10 min at room temperature. Chromogenic detection was conducted by incubating tissue sections with DAB for 10 min. Counterstaining was conducted by incubating the tissue sections with hematoxylin for 5 min at room temperature. DAB and hematoxylin were contained in the BOND IHC Polymer Detection Kit (cat. no. DS9800). The IHC staining was quantified using a scoring system described previously (24). The percentage of positive cells was scored as follows: i) Score of 1, ≤24%; ii) score of 2, 25-50%; iii) score of 3, 51-75%; and iv) score of 4, ≥76%. The intensity of IHC staining was scored as follows: i) Score of 0, negative; ii) score of 1, weak; iii) score of 2, moderate; and (iv) score of 3, strong. The total score was calculated as follows: Total score = positive percentage score x intensity score.

For Safranin O-Fast Green and H&E staining, the Safranin O-Fast Green staining kit (cat. no. PH1852; Phygene) and the H&E staining kit (cat. no. PH0516; Phygene) were loaded into BOND-III Automated IHC Stainer. For Safranin O-Fast Green staining, after deparaffinization and rehydration as aforementioned, the tissue sections were incubated with Safranin O solution for 10 min at the room temperature. Next, the sections were washed with 70% ethanol for 30 sec. The sections were then incubated with Fast Green solution for 5 min. After the staining, the sections were washed with graded alcohols (80, 95 and 100%) for 30 sec each. Finally, the sections were washed with xylene (2X) for 1 min. For H&E saining, after deparaffinization and rehydration, the tissue sections were incubated hematoxylin solution for 10 min at the room temperature and then washed with bluing reagent for 5 min. Next, the sections were incubated with eosin solution for 5 min. Washing with graded alcohols and with xylene was as aforementioned. The staining images were captured using a DM4000 B LED microscope (Leica Microsystems, Inc.) at x20 or x40 magnification.

Conventional western blotting was performed as previously described (25). Briefly, total proteins were extracted from the PMCs using RIPA lysis buffer (Beyotime Institute of Biotechnology) and protein concentration was determined using a BCA kit (cat. no. P0012; Beyotime Institute of Biotechnology). The proteins (25 µg/lane) were separated on 10% gels using SDS-PAGE, transferred onto nitrocellulose membranes, blocked using 5% BSA (Gibco; Thermo Fisher Scientific, Inc.) in TBST (0.1% Tween-20) at 37˚C for 30 min and then incubated with primary antibodies at 4˚C overnight. The following antibodies were used: Anti-NLRP3 (1:1,000; cat. no. 19771-1-AP; Wuhan Sanying Biotechnology), anti-GSDMD-NT (1:1,000; cat. no. 36425; Cell Signaling Technology, Inc.), anti-cleaved caspase-1 (1:1,000; cat. no. 89332; Cell Signaling Technology, Inc.), anti-mature IL-1β (1:1,000; cat. no. 83186; Cell Signaling Technology, Inc.) and anti-β-actin (1:2,000; cat. no. 20536-1-AP; Wuhan Sanying Biotechnology). The membranes were washed with 1X TBST three times (5 min/time) and incubated with HRP-conjugated secondary antibodies (1:5,000; cat. no. SA00001-2; Wuhan Sanying Biotechnology) at room temperature for 1 h. The protein bands were visualized using an enhanced chemiluminescence kit (BeyoECL Star; Beyotime Institute of Biotechnology) and ChemiScope 6200T imager (Clinx Science Instruments Co., Ltd.). The intensities of protein bands were quantitated using ImageJ (v1.5.4; National Institutes of Health) based on three biological repeats.

Data are presented as the mean ± SD. Statistical analysis was conducted using GraphPad Prism 8.10 (GraphPad Software; Dotmatics). The Wilcoxon's signed-rank test was performed for Sham vs. DMM comparisons due to the paired nature of these two groups, whilst the Wilcoxon's rank-sum test was performed for DMM vs. DMM + GCK comparisons due to the unpaired nature of these two groups. Bonferroni's correction was conducted on all P-values yielded by these two aforementioned tests. Since two tests were performed within each group, P<0.025 was considered to indicate a statistically significant difference in these two cases. Either Kruskal-Wallis test followed by Dunn's test (staining scores) or one-way ANOVA followed by Tukey's post hoc test (numerical data) was used for the rest of the multiple comparisons in Figs. 1 and 3. P<0.05 was considered to indicate a statistically significant difference.

To check the influence of GCK on the viability of PMCs, cells were cultured in the presence of different concentrations of GCK (0, 10, 20, 50, 100 and 150 µM) for 48 h. CCK-8 assay results revealed that GCK conferred no cytotoxicity towards PMCs, even at the concentration of 150 µM (Fig. 1A). Furthermore, concentrations of 20 and 50 µM GCK significantly increased cell viability compared with that in the control cells (Fig. 1A). A total of 10 and 50 GCK µM were used in the following studies to assess whether GCK had dose-dependent effects on PMCs.

In addition, the viability of PMCs was found to be significantly reduced by TNF-α treatment. However, co-treatment with GCK (20 and 50 µM) significantly restored their cell viability (Fig. 1B). To assess the influence of GCK on TNF-α-induced inflammation and ECM dysregulation, the secretion of ECM regulatory factors were measured using ELISA. The results indicated that TNF-α induced a significant increase in IL-6, MMP13, ADAMTS5 and MMP3 levels (Fig. 1C-F), whilst significantly decreasing aggrecan and collagen II levels (Fig. 1G and H). GCK appeared to have reversed these alterations in a dose-dependent manner (Fig. 1C-H). Immunofluorescence staining also revealed that GCK partially suppressed the MMP13 expression that was induced by TNF-α whilst rescuing the collagen II expression that was reduced by TNF-α in PMCs (Fig. 1I and J).

To explore the potential effects of GCK on osteoarthritis in vivo, an osteoarthritis model was established in mice by DMM surgery, without significant adverse effects among the animals, such as significant loss of body weight and consistent bleeding. Knee joints were collected, paraffin-embedded, sectioned and stained. Safranin O-Fast Green and H&E staining revealed that DMM surgery resulted in cartilage erosion in the femur and tibia, loss of the superficial zone and reduced uncalcified cartilage (red and black arrows; .. 2A and C). However, these pathological changes were mitigated by GCK supplementation (Fig. 2A and C).

To quantify the aforementioned changes, the sections were scored using the Osteoarthritis Research Society International (OARSI) semi-quantitative grading system, which assesses the lesion severity and the affected area in both the femur and tibia (26). The DMM group had a significantly increased OARSI score compared with that in the sham group (Fig. 2B). By contrast, the DMM mice treated with GCK had a significantly lower OARSI score compared with that in the DMM-only group (Fig. 2B). Protein expression was then assessed using immunohistochemistry. The mice treated with GCK exhibited significantly increased collagen II (Fig. 2E and D) and significantly decreased MMP13 expression (Fig. 2G and F) compared with that in the DMM-only group.

In PMCs, TNF-α treatment stimulated a significant increase in the levels of NLRP3, GSDMD-NT, cleaved caspase-1 and mature IL-1β, the markers indicating the occurrence of pyroptosis (5) (Fig. 3A and C-F). However, GCK treatment suppressed the increase of these proteins in a dose-dependent manner (Fig. 3A and C-F). TUNEL assay was then used to examine the extent of cell death of PMCs. The TNF-α treatment group had a significantly higher level of TUNEL-positive cells compared with that in the DMSO group (Fig. 3B and G). However, GCK treatment decreased the percentage of TUNEL-positive PMCs induced by TNF-α (Fig. 3B and G).

IHC analysis of the knee joint tissue sections from DMM mice with or without GCK treatment confirmed that NLRP3 and GSDMD-NT expression was significantly increased in the cartilage of the DMM mice compared with that in the sham control group (Fig. 4A-C). Compared with that in the DMM mice without GCK treatment, the DMM mice that received GCK treatment exhibited significantly reduced NLRP3 and GSDMD-NT expression (Fig. 4A-C).